Petar Milin
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Petar Milin has a total of 74 co-authors. Here are the top co-authors by number of shared publications:
- Dagmar Divjak
University of Birmingham
16 shared publications - R. Harald Baayen
University of Tübingen, Tübingen, Germany
8 shared publications - Dušica Filipović Đurđević
Faculty of Philosophy - University of Belgrade
6 shared publications Dagmar Divjak
Department of Modern Languages, University of Birmingham, Birmingham, UK
5 shared publicationsMichael Ramscar
Department of Linguistics, University of Tübingen
5 shared publications- Filip Nenadić
University of Alberta
4 shared publications
Latest Publications
- Open Access
- 3 September 2026
- Published by Taylor & Francis in Journal of Applied Statistics
- p. 1-34
The analysis of fixed and random effects in a log-linear modeling framework
1. This paper addresses a significant methodological and statistical challenge: the absence of a universally accepted approach to mixed effects in log-linear models, a powerful type of generalized ...
- 3 September 2026
- Published by Taylor & Francis in Journal of Applied Statistics
- p. 1-34
Open AccessThe analysis of fixed and random effects in a log-linear modeling framework
1. This paper addresses a significant methodological and statistical challenge: the absence of a universally accepted approach to mixed effects in log-linear models, a powerful type of generalized ...
- 25 August 2026
- Published by Springer Nature
- p. 1-9
Eye-Tracking in Online Research
- Open Access
- 1 August 2026
- Published by American Psychological Association (APA) in Journal of Experimental Psychology: Learning, Memory, and Cognition
- Vol. 52 (8) , 1489-1509
The games words and contexts play: How readers navigate syntagmatic and paradigmatic relations.
While previous studies have focused on how contextual elements influence reading, our research adopts a novel approach to the dynamic interaction between words and their context. Using eye movements during naturalistic Serbian sentence reading, we analyzed how both morphological and semantic information carried by consecutive nouns in a sentence impact reading behavior. We considered a broad range of potential predictors, categorized as either benchmark (word length and frequency), paradigmatic (relative entropy, letter-triplet activation, and competition), or syntagmatic (cosine similarity, contextual similarity, and typicality) to predict four eye movement measures: landing position, first fixation duration, total fixation duration, and average pupil size. A multivariate statistical model that enabled simultaneous testing of multiple predictors against several dependent variables revealed a complex interaction among word properties within context. While previously identified predictors such as length and frequency come out strong, our approach reveals the relational nature of word- and context-specific effects and brings out how both paradigmatic and syntagmatic information shapes language processing. (PsycInfo Database Record (c) 2025 APA, all rights reserved).
- 1 August 2026
- Published by American Psychological Association (APA) in Journal of Experimental Psychology: Learning, Memory, and Cognition
- Vol. 52 (8) , 1489-1509
Open AccessThe games words and contexts play: How readers navigate syntagmatic and paradigmatic relations.
While previous studies have focused on how contextual elements influence reading, our research adopts a novel approach to the dynamic interaction between words and their context. Using eye movements during naturalistic Serbian sentence reading, we analyzed how both morphological and semantic information carried by consecutive nouns in a sentence impact reading behavior. We considered a broad range of potential predictors, categorized as either benchmark (word length and frequency), paradigmatic (relative entropy, letter-triplet activation, and competition), or syntagmatic (cosine similarity, contextual similarity, and typicality) to predict four eye movement measures: landing position, first fixation duration, total fixation duration, and average pupil size. A multivariate statistical model that enabled simultaneous testing of multiple predictors against several dependent variables revealed a complex interaction among word properties within context. While previously identified predictors such as length and frequency come out strong, our approach reveals the relational nature of word- and context-specific effects and brings out how both paradigmatic and syntagmatic information shapes language processing. (PsycInfo Database Record (c) 2025 APA, all rights reserved).
- Open Access
- 9 March 2026
- Published by Walter de Gruyter GmbH in Linguistics Vanguard
- Vol. 12 (s1) , 187-195
The synergy of linguistic theories and models of learning
Emergence is a key concept in usage-based linguistic theories, which view natural languages as complex systems learned through exposure to usage, by relying on general cognitive functions rather than an innate language faculty and its underlying universal grammar posited by generativist theories. Despite its importance, however, the emergence of language knowledge has often been treated axiomatically rather than being empirically substantiated. We argue that learning, understood as a gradual change in knowledge, can serve as an operational definition of emergence of knowledge through usage. Integrating usage-based linguistics with the psychology of learning creates a productive synergy. For linguistics, a focus on learning processes constrains theoretical abstractions to those learnable from data. For psychology, language provides a unique and challenging testbed for understanding real-world learning mechanisms. We illustrate these points with case studies on the tense-aspect systems of English and Polish, shedding new light on the role of imperfect contingencies in shaping linguistic patterns while testing psychological models of learning.
- 9 March 2026
- Published by Walter de Gruyter GmbH in Linguistics Vanguard
- Vol. 12 (s1) , 187-195
Open AccessThe synergy of linguistic theories and models of learning
Emergence is a key concept in usage-based linguistic theories, which view natural languages as complex systems learned through exposure to usage, by relying on general cognitive functions rather than an innate language faculty and its underlying universal grammar posited by generativist theories. Despite its importance, however, the emergence of language knowledge has often been treated axiomatically rather than being empirically substantiated. We argue that learning, understood as a gradual change in knowledge, can serve as an operational definition of emergence of knowledge through usage. Integrating usage-based linguistics with the psychology of learning creates a productive synergy. For linguistics, a focus on learning processes constrains theoretical abstractions to those learnable from data. For psychology, language provides a unique and challenging testbed for understanding real-world learning mechanisms. We illustrate these points with case studies on the tense-aspect systems of English and Polish, shedding new light on the role of imperfect contingencies in shaping linguistic patterns while testing psychological models of learning.
- Open Access
- 8 September 2025
- Published by Wiley in Language Learning
- Vol. 76 (2) , 494-527
Error Correction Learning of Second Language Verbal Morphology: Associating Imperfect Contingencies in Naturalistic Frequency Distributions
We investigate what is learned from exposure to usage in verbal morphology using an error correction mechanism within an associative learning framework. We computationally simulated how second language (L2) learners would respond to naturalistic input of aspectual usage, characterized by “imperfect contingencies,” given two types of instructions: grammatical versus lexical. To test these predictions, English native speakers ( N = 80) completed three online training sessions in two conditions (grammatical vs. lexical) over 3 days, learning 21 Polish verbs across 189 exposures; and a 63‐item posttest on Day 4 (50% seen, 50% grammatical). The results confirmed the simulation predictions: The grammatical group performed better through stronger performance in contexts that allow only one aspect; the lexical group was slightly better in contexts where both aspects were possible. Rules offer some advantage early on, especially when the exemplars are already unambiguous, whereas an exemplar‐based approach promises a more flexible system in the longer run.
- 8 September 2025
- Published by Wiley in Language Learning
- Vol. 76 (2) , 494-527
Open AccessError Correction Learning of Second Language Verbal Morphology: Associating Imperfect Contingencies in Naturalistic Frequency Distributions
We investigate what is learned from exposure to usage in verbal morphology using an error correction mechanism within an associative learning framework. We computationally simulated how second language (L2) learners would respond to naturalistic input of aspectual usage, characterized by “imperfect contingencies,” given two types of instructions: grammatical versus lexical. To test these predictions, English native speakers ( N = 80) completed three online training sessions in two conditions (grammatical vs. lexical) over 3 days, learning 21 Polish verbs across 189 exposures; and a 63‐item posttest on Day 4 (50% seen, 50% grammatical). The results confirmed the simulation predictions: The grammatical group performed better through stronger performance in contexts that allow only one aspect; the lexical group was slightly better in contexts where both aspects were possible. Rules offer some advantage early on, especially when the exemplars are already unambiguous, whereas an exemplar‐based approach promises a more flexible system in the longer run.
- Open Access
- 8 June 2025
- Published by Springer Nature in Journal of Cultural Cognitive Science
- Vol. 9 (2) , 243-254
The interplay of sensory and motoric information on processing emotions in narrative texts
This study examines the comprehension of characters’ emotional states in written narrative texts. Recent theories of embodied cognition suggest that emotions are grounded in somatosensory information involving ‘reexperiencing’ of actions and perceptions. In addition, views of graded embodied cognition suggest that there are levels of the embodiment of language, focusing on how perceptual and motor information interact during text comprehension. In the present experiment, one-paragraph stories were written to express six basic emotions: fear, sadness, anger, disgust, happiness, and surprise. Four stories were written for each emotion and for each story four critical sentences were composed. The critical sentences comprised a combination of emotion- and action-based components that matched or mismatched the story. Participants read the stories and the critical sentences. Their task was to respond to a question about the emotional state of the story’s main character, while we measured response latencies and errors. The results of the error rate analyses suggest that, while reading texts, sensory knowledge about characters’ emotional states is activated, but this knowledge is significantly moderated by action-based knowledge. A computational model was used to further confirm these results. The model was trained to predict emotion and action words from critical sentences using linguistic context from stories. Both, the action and emotional words activations showed a distinct effect on participants’ comprehension accuracy.
- 8 June 2025
- Published by Springer Nature in Journal of Cultural Cognitive Science
- Vol. 9 (2) , 243-254
Open AccessThe interplay of sensory and motoric information on processing emotions in narrative texts
This study examines the comprehension of characters’ emotional states in written narrative texts. Recent theories of embodied cognition suggest that emotions are grounded in somatosensory information involving ‘reexperiencing’ of actions and perceptions. In addition, views of graded embodied cognition suggest that there are levels of the embodiment of language, focusing on how perceptual and motor information interact during text comprehension. In the present experiment, one-paragraph stories were written to express six basic emotions: fear, sadness, anger, disgust, happiness, and surprise. Four stories were written for each emotion and for each story four critical sentences were composed. The critical sentences comprised a combination of emotion- and action-based components that matched or mismatched the story. Participants read the stories and the critical sentences. Their task was to respond to a question about the emotional state of the story’s main character, while we measured response latencies and errors. The results of the error rate analyses suggest that, while reading texts, sensory knowledge about characters’ emotional states is activated, but this knowledge is significantly moderated by action-based knowledge. A computational model was used to further confirm these results. The model was trained to predict emotion and action words from critical sentences using linguistic context from stories. Both, the action and emotional words activations showed a distinct effect on participants’ comprehension accuracy.
- Open Access
- 28 April 2025
- Published by John Benjamins Publishing Company in Linguistic Approaches to Bilingualism
- Vol. 15 (4) , 487-517
Optimising participant grouping methods in bilingualism studies
This research addresses two major challenges in studying second language acquisition and bilingualism: reducing overlap in predictor variables and correctly classifying participants into language proficiency levels. Too many relevant predictors can harm statistical analysis due to an increased chance of overlap, known as multicollinearity. To tackle this, we use Principal Component Analysis (PCA) on selected predictors to identify proficiency indicators, combining the length of stay in the UK and language test scores. Additionally, traditional methods, especially IELTS-based proficiency classifications, often miss subtle differences in language skills, particularly when they fail to consider how long participants have been exposed to the target language. We counter this by using non-hierarchical Cluster Analysis (NCA) for a grounded, data-driven way of detecting distinct language proficiency groups. This new approach is demonstrated on a dataset of eye movements from reading tasks, collected from Chinese–English bilinguals in the UK.
- 28 April 2025
- Published by John Benjamins Publishing Company in Linguistic Approaches to Bilingualism
- Vol. 15 (4) , 487-517
Open AccessOptimising participant grouping methods in bilingualism studies
This research addresses two major challenges in studying second language acquisition and bilingualism: reducing overlap in predictor variables and correctly classifying participants into language proficiency levels. Too many relevant predictors can harm statistical analysis due to an increased chance of overlap, known as multicollinearity. To tackle this, we use Principal Component Analysis (PCA) on selected predictors to identify proficiency indicators, combining the length of stay in the UK and language test scores. Additionally, traditional methods, especially IELTS-based proficiency classifications, often miss subtle differences in language skills, particularly when they fail to consider how long participants have been exposed to the target language. We counter this by using non-hierarchical Cluster Analysis (NCA) for a grounded, data-driven way of detecting distinct language proficiency groups. This new approach is demonstrated on a dataset of eye movements from reading tasks, collected from Chinese–English bilinguals in the UK.
- Open Access
- 14 March 2025
- Published by Walter de Gruyter GmbH in Corpus Linguistics and Linguistic Theory
- Vol. 22 (2) , 531-561
On the learnability of aspectual usage
We examine the learnability of grammatical aspect in Polish through the application of an error-correction learning algorithm to corpus data. We explore whether mastery of aspectual usage can be attained solely through exposure to co-occurrence patterns, or if an understanding of the abstract semantic distinctions discussed in the aspectual literature are a prerequisite. To validate the model, we compare the corpus-based findings with data on aspectual usage from a survey of Polish L1 users. Our findings show that aspectual preferences are best accounted for by a model that relies on co-occurrence information to predict a particular lemma, without reference to aspectual information. The models incorporating aspectual information, either by predicting aspect from co-occurrence information or from abstract semantic labels, performed adequately but were only able to capture usage preferences relating to one of the two aspects. This study contributes insights into the role of usage in the acquisition of linguistic categories and underscores the importance of integrating diverse evidence sources and methodologies for the development and validation of linguistic theories.
- 14 March 2025
- Published by Walter de Gruyter GmbH in Corpus Linguistics and Linguistic Theory
- Vol. 22 (2) , 531-561
Open AccessOn the learnability of aspectual usage
We examine the learnability of grammatical aspect in Polish through the application of an error-correction learning algorithm to corpus data. We explore whether mastery of aspectual usage can be attained solely through exposure to co-occurrence patterns, or if an understanding of the abstract semantic distinctions discussed in the aspectual literature are a prerequisite. To validate the model, we compare the corpus-based findings with data on aspectual usage from a survey of Polish L1 users. Our findings show that aspectual preferences are best accounted for by a model that relies on co-occurrence information to predict a particular lemma, without reference to aspectual information. The models incorporating aspectual information, either by predicting aspect from co-occurrence information or from abstract semantic labels, performed adequately but were only able to capture usage preferences relating to one of the two aspects. This study contributes insights into the role of usage in the acquisition of linguistic categories and underscores the importance of integrating diverse evidence sources and methodologies for the development and validation of linguistic theories.
- Open Access
- 11 December 2024
- Published by Elsevier in Research Methods in Applied Linguistics
- Vol. 4 (1)
Introducing fluency measures to the elicited imitation task
- 11 December 2024
- Published by Elsevier in Research Methods in Applied Linguistics
- Vol. 4 (1)
Open AccessIntroducing fluency measures to the elicited imitation task
- Open Access
- 17 September 2024
- Published by Wiley in Language Learning
- Vol. 75 (3) , 623-665
Order Effects in Second Language Learning
We explore how general principles of learning apply to and combine with usage‐based approaches to language learning and teaching, with a focus on the effects of order of exposure to new information in second language (L2) instruction. Although the effects of input spacing and timing on memory and learning have been previously explored (see Rogers, 2020; Shintani, 2017, for an overview), the effects of order of exposure to the input remain understudied. In this study, we tested whether order of exposure to information in the L2 plays a role in L2 learning and whether an optimal order where the most reliable cues are introduced first ensures that foreign language learners are better equipped to form representations that allow flexible, yet accurate generalisations. We analysed data collected through a training study that teaches the English article system to first language (L1) speakers whose language does not have markers of definiteness.
- 17 September 2024
- Published by Wiley in Language Learning
- Vol. 75 (3) , 623-665
Open AccessOrder Effects in Second Language Learning
We explore how general principles of learning apply to and combine with usage‐based approaches to language learning and teaching, with a focus on the effects of order of exposure to new information in second language (L2) instruction. Although the effects of input spacing and timing on memory and learning have been previously explored (see Rogers, 2020; Shintani, 2017, for an overview), the effects of order of exposure to the input remain understudied. In this study, we tested whether order of exposure to information in the L2 plays a role in L2 learning and whether an optimal order where the most reliable cues are introduced first ensures that foreign language learners are better equipped to form representations that allow flexible, yet accurate generalisations. We analysed data collected through a training study that teaches the English article system to first language (L1) speakers whose language does not have markers of definiteness.