PosterPDF Available

Time in language: past and future are differently processed and both verbs and adverbs matter

Authors:

Abstract

The main aim of this work is to investigate the processing of past and future time reference during language comprehension. Mentally locating events in the past or future plays a key role in many fundamental aspects of human cognition and behavior, such as planning, decision-making, and self-regulation. From the psycholinguistic perspective, previous studies have mainly focused on the processing of past and present time reference, while the comparison between past and future is still deficient. Moreover, previous studies have mainly focused on the processing of past and non-past time reference violations related to an inflected verb presented within a sentence. In this study, we fill a clear gap in the literature by investigating both the processing of past/future verbs presented in isolation and within sentences and by investigating the role played by different types of temporal adverbs.
Time in language: past and future are differently processed and both verbs and adverbs matter
Nicoletta Biondo1, Marielena Soilemezidi2 & Simona Mancini3
nicoletta.biondo@unisi.it
Introduction
PAst DIscourse LInking Hypothesis (PADILIH): Past is more complex than non-
past (present/future) processing since only past time reference requires a link
to a time specified by the discourse context [1]. The past-present dissociation
was found in numerous studies. The past-future dissociation is still unclear.
Past and future both imply a temporal displacement of an event outside the
“now”, but they also differ: the past refers to inalterable events; the future to
alterable events, to possible worlds [2][3].
Question 1: Are past and future differently processed during language
processing?
In many languages (e.g., Spanish), past and future information is encoded in
verb morphology and, occasionally, through temporal adverbs. No study has
disentangled the role played by verb and adverbs.
Question 2: What is the role of verbs and adverbs during time processing?
-Lorem ipsum dolor sit amet,consectetur adipiscing elit. Morbi ligula.
Etiam in diam. Nullam libero risus, malesuada et, adipiscing
eget, tempor sit amet, eros.
-Lorem ipsum dolor sit amet, consectetur adipiscing elit. Morbi
ligula. Etiam in diam. Nullam libero risus, malesuada et,
adipiscing eget, tempor sit amet, eros.
Method
EYE-TRACKING WHILE-READING STUDY (N=60)
Question 1: Time Reference (TR: past, future) x Verb Match (VM: match, mismatch)
e.g. Gracias a la beca (Thanks to the scholarship), …
PAST el año pasado los investigadores progresaron / *progresarán
FUTURE el próximo año los investigadores progresarán / *progresaron..
(last year/next year the researchers made/will make progress)
en sus estudios sobre la polución (on their studies on pollution).
Question 2: Adverb Type (AT: deictic, non deictic) x Time Reference (TR: past, future)
e.g. Gracias a la beca (Thanks to the scholarship), …
DEICTIC ... el año pasado/el próximo año los investigadores progresaron / progresarán..
NON-DEICTIC ... a lo largo del año los investigadores progresaron / progresarán
(last-next year/throughout the year the researchers made/will make progress)
... en sus estudios sobre la polución (on their studies on pollution).
TEMPORAL DECISION TASK (N=60)
Question 2: Time Reference (TR: past, future)
Results
Discussion
In the ET study, past mismatches yielded early effects while future mismatch effects emerged only in total time, in line with [1].
Past and future time reference violations are differently treated by the parser [1].
Past took longer than future categorization in isolated verbs (TDT), in line with [1]. However, when verbs were presented within
(grammatical non-/deictic) sentences (ET), future triggered longer go-past and total reading times compared to past, contra [1].
Past and future time processing on the verb differs if verbs are presented in isolation or within a sentence.
This difference cannot be due to discourse-linking [1]. The abstract/alterable representation of future events [2][3] may
slowdown future processing in a sentential context.
References
[1] Bastiaanse, R., Bamyaci, E., Hsu, C. J., Lee, J., Duman, T. Y., & Thompson, C. K. (2011).
Journal of Neurolinguistics, 24(6), 652-673.
[2] Comrie, B. (1985). Cambridge university press.
[3] Bochnak, M. R. (2019). Language and Linguistics Compass, 13(1), e12307.
1 2 3
progresaron
TEMPORAL DECISION TASK
EYE-TRACKING (ANALYSIS 1)
EYE-TRACKING (ANALYSIS 2)
Parsimonious LMEM analyses
Sum contrasts (past, match/deictic -0.5)
Bonferroni correction (α = 0.0045)
Higher regression probabilities to the adverb were found
in the presence of mismatching verbs or when (non-
deictic) adverbs did not provide a specific time frame.
→ Temporal adverbs do not facilitate time processing on
the verb but they play a role in reanalysis routines.
TR x VM TR x VM VM VM VM VM VM VM VM
TR TR AT
TR
ResearchGate has not been able to resolve any citations for this publication.
  • R Bastiaanse
  • E Bamyaci
  • C J Hsu
  • J Lee
  • T Y Duman
  • C K Thompson
Bastiaanse, R., Bamyaci, E., Hsu, C. J., Lee, J., Duman, T. Y., & Thompson, C. K. (2011). Journal of Neurolinguistics, 24(6), 652-673.
Language and Linguistics Compass
  • M R Bochnak
Bochnak, M. R. (2019). Language and Linguistics Compass, 13(1), e12307.
ANALYSIS 2) • Parsimonious LMEM analyses • Sum contrasts (past, match/deictic -0.5) • Bonferroni correction
  • Eye-Tracking
EYE-TRACKING (ANALYSIS 2) • Parsimonious LMEM analyses • Sum contrasts (past, match/deictic -0.5) • Bonferroni correction (α = 0.0045)