In this chapter, we highlight Bill VanPatten’s numerous contributions to the fields of second language acquisition (SLA) and instructed second language acquisition (ISLA), and to the language teaching profession. We begin with an overview of his major professional achievements, followed by an exploration of his early work and the origins of his model of input processing. We underscore the significant contribution that his model of input processing makes to theory building in SLA and subsequently discuss his pioneering instructional innovation, processing instruction, along with the prolific body of research it has inspired. After briefly noting the importance of his work on acquisition and processing more broadly, we conclude with an overview of the individual chapters included in the volume.
This self-paced reading study investigated the processing of case marking with the particle a in non-native Spanish. Experiment 1 tested online sensitivity to case marking on direct and indirect objects. Unlike native speakers, intermediate L2 learners showed effects only with indirect objects. For Experiment 2, direct objects were doubled with a preverbal clitic, to see if the addition was associated with more native-like processing, as has been shown in previous research with very advanced L2 learners (Jegerski, 2015). Again, online sensitivity to case marking was only observed with the ditransitive stimuli. Thus, clitic doubling only seems to make a difference at sufficiently high levels of proficiency, perhaps because of increasingly robust representations of grammatical forms in the developing L2 system.
A vital part of understanding language is linking referring expressions to the appropriate antecedents. In null subject languages like Spanish, person-number inflections on verbs are essential to ensuring the appropriate connections are made. However, second language (L2) learners have limited processing capabilities (VanPatten, 1996, 2015), and may rely on strategies that steer them away from inflections. The current study examines whether first language English/L2 Spanish learners process verb inflections that signal subject shifts or if they prefer to process lexical items (i.e., overt subjects) as predicted by VanPatten’s lexical preference principle. The results of this study indicate that non-natives do not rely on inflections to link antecedents and referring expressions, and that L2 processing is facilitated by overt subject pronouns.
This chapter investigates the role of first language (L1) transfer, second language (L2) proficiency, and working memory (WM) on processing subject-verb number agreement in Spanish. Intermediate and advanced English and Romanian learners of Spanish and Spanish monolinguals completed a WM test and a reading eye-tracking task with sentences with adjacent subject-verb number agreement/disagreement. Results revealed that all groups were sensitive to violations but processed them differently, depending on their L1 (the less morphologically rich, the more reliance on subjects), proficiency (the more advanced, the more reliance on verbs), and WM (the higher the span, the more sensitivity to violations). These findings indicate that higher L1/L2 similarity, L2 proficiency, and WM facilitate L2 verbal morphosyntactic processing.
VanPatten’s (1990) investigation of attention to form and meaning during input processing remains among the most cited and replicated studies in second language research. Early replications (Barcroft, 2001; Greenslade, Bouden, & Sanz, 1999; Wong, 2001) kept materials intact, changing modality or language, yielding comparable results. Later replications operationalized comprehension differently, changing text and assessment, and concluded even low proficiency learners can consciously process form, regardless of salience or meaning, without detriment to comprehension (Leow, Hsieh, & Moreno, 2008; Morgan-Short, Heil, Botero-Moriarty, & Ebert, 2012; Morgan-Short et al., 2018). This chapter compares these texts and assessments. Results suggest the new assessment obscures differences between proficiencies and modalities, and the new, more complex text hides task effects identified in the original strand.
This study investigates the contextual constraint principle in input processing, which states that learners may rely less on a first-noun-as-subject/agent strategy if preceding context constrains the interpretation of a clause or sentence. Second language learners of Spanish (N = 39) completed a sentence interpretation task in which they heard OVS sentences in various context conditions: neutral versus constraining context appearing prior to and after OVS clauses. Results revealed greater accuracy on OVS sentence interpretation when constraining context was available, particularly when it appeared prior to the target OVS clause. The results suggest that first noun reliance is significantly weakened when constraining context is available, and that the contextual constraint principle should be expanded to include context in post target-clause position.
Research on input processing (IP) to date has focused largely on sentences and acquisition of morphosyntax. Other research has addressed IP at the lexical level for different aspects of vocabulary acquisition. This chapter clarifies how understanding multilevel IP can improve our understanding of IP across the board. Section 1 defines multilevel IP and clarifies its overarching role when learners are exposed to input. Section 2 summarizes research on sentence-level IP and identifies areas in which research on lexical IP can be informative. Section 3 summarizes research on lexical IP, noting areas in which research on sentence-level IP can be informative. Finally, Section 4 discusses challenges to be met in new efforts to investigate and advance theory related to multilevel IP.
Trials-to-criterion – a method commonly used in applied behavior analysis as an alternative to the pre-/post-test study design – has been used to observe and assess input processing in processing instruction (PI) research. The results from trials-to-criterion studies have made important contributions to our understanding of the effects of PI in input processing, particularly regarding the role of explicit information. In order to promote trials-to-criterion as a method in PI research, this chapter reviews and discusses studies that have used trials-to-criterion data to investigate the effects of PI on second language processing. It also examines crucial design choices in the implementation of trials-to-criterion as a measure of language processing. Finally, it discusses future considerations for trials-to-criterion research.
We examined whether French L2 learners trained with processing instruction (PI) or traditional instruction (TI) on the French causative, would transfer training effects to the passive construction. Experiment 2 included explicit information (EI) but not Experiment 1. Accuracy results and eye movement patterns showed PI had a positive impact on participants’ processing of the causative, but neither treatment had any effect on the passive. The null effect could be attributed to the low number of tokens used and to the nature of the passive structure. Furthermore, PI training on the causative might have trained learners to rely on the necessary cues to process causative sentences correctly, but this might not have been sufficient to decrease reliance on the first noun principle.
Research on processing instruction (PI) has shown that PI is an effective instructional intervention for learners of various languages and contexts. This chapter addresses a basic question: Why does PI work? In the first section, I provide an overview of the language faculty and discuss how PI differs from other research in terms of the nature of language. The second section explores how VanPatten’s model of input processing (VanPatten, 1996, 2004, 2020) can be understood within larger frameworks that integrate language processing and acquisition. The third section discusses the characteristics of PI and the findings of PI research within the wider context of growth within the language faculty. Finally, I conclude with thoughts on future directions of PI research.
Evidence for the input processing (IP) model (VanPatten, 1996, 2004, 2007, 2015a, 2020) and the benefits of processing instruction (PI) comes largely from offline tasks, despite the fact that both are concerned with online sentence comprehension. This chapter aims to stimulate more online research in these areas. I first provide a brief summary of techniques that measure real-time sentence comprehension and then offer suggestions for how they can be used in research on the IP model and PI. Lastly, I review recent online studies of the effects of PI on second language (L2) learners’ real-time sentence comprehension. I end by proposing two new hypotheses regarding the potential that PI has to benefit L2 learners’ online processing of input.
In this chapter, we highlight Bill VanPatten’s numerous contributions to the fields of second language acquisition (SLA) and instructed second language acquisition (ISLA), and to the language teaching profession. We begin with an overview of his major professional achievements, followed by an exploration of his early work and the origins of his model of input processing. We underscore the significant contribution that his model of input processing makes to theory building in SLA and subsequently discuss his pioneering instructional innovation, processing instruction, along with the prolific body of research it has inspired. After briefly noting the importance of his work on acquisition and processing more broadly, we conclude with an overview of the individual chapters included in the volume.
This self-paced reading study investigated the processing of case marking with the particle a in non-native Spanish. Experiment 1 tested online sensitivity to case marking on direct and indirect objects. Unlike native speakers, intermediate L2 learners showed effects only with indirect objects. For Experiment 2, direct objects were doubled with a preverbal clitic, to see if the addition was associated with more native-like processing, as has been shown in previous research with very advanced L2 learners (Jegerski, 2015). Again, online sensitivity to case marking was only observed with the ditransitive stimuli. Thus, clitic doubling only seems to make a difference at sufficiently high levels of proficiency, perhaps because of increasingly robust representations of grammatical forms in the developing L2 system.
A vital part of understanding language is linking referring expressions to the appropriate antecedents. In null subject languages like Spanish, person-number inflections on verbs are essential to ensuring the appropriate connections are made. However, second language (L2) learners have limited processing capabilities (VanPatten, 1996, 2015), and may rely on strategies that steer them away from inflections. The current study examines whether first language English/L2 Spanish learners process verb inflections that signal subject shifts or if they prefer to process lexical items (i.e., overt subjects) as predicted by VanPatten’s lexical preference principle. The results of this study indicate that non-natives do not rely on inflections to link antecedents and referring expressions, and that L2 processing is facilitated by overt subject pronouns.
This chapter investigates the role of first language (L1) transfer, second language (L2) proficiency, and working memory (WM) on processing subject-verb number agreement in Spanish. Intermediate and advanced English and Romanian learners of Spanish and Spanish monolinguals completed a WM test and a reading eye-tracking task with sentences with adjacent subject-verb number agreement/disagreement. Results revealed that all groups were sensitive to violations but processed them differently, depending on their L1 (the less morphologically rich, the more reliance on subjects), proficiency (the more advanced, the more reliance on verbs), and WM (the higher the span, the more sensitivity to violations). These findings indicate that higher L1/L2 similarity, L2 proficiency, and WM facilitate L2 verbal morphosyntactic processing.
VanPatten’s (1990) investigation of attention to form and meaning during input processing remains among the most cited and replicated studies in second language research. Early replications (Barcroft, 2001; Greenslade, Bouden, & Sanz, 1999; Wong, 2001) kept materials intact, changing modality or language, yielding comparable results. Later replications operationalized comprehension differently, changing text and assessment, and concluded even low proficiency learners can consciously process form, regardless of salience or meaning, without detriment to comprehension (Leow, Hsieh, & Moreno, 2008; Morgan-Short, Heil, Botero-Moriarty, & Ebert, 2012; Morgan-Short et al., 2018). This chapter compares these texts and assessments. Results suggest the new assessment obscures differences between proficiencies and modalities, and the new, more complex text hides task effects identified in the original strand.
This study investigates the contextual constraint principle in input processing, which states that learners may rely less on a first-noun-as-subject/agent strategy if preceding context constrains the interpretation of a clause or sentence. Second language learners of Spanish (N = 39) completed a sentence interpretation task in which they heard OVS sentences in various context conditions: neutral versus constraining context appearing prior to and after OVS clauses. Results revealed greater accuracy on OVS sentence interpretation when constraining context was available, particularly when it appeared prior to the target OVS clause. The results suggest that first noun reliance is significantly weakened when constraining context is available, and that the contextual constraint principle should be expanded to include context in post target-clause position.
Research on input processing (IP) to date has focused largely on sentences and acquisition of morphosyntax. Other research has addressed IP at the lexical level for different aspects of vocabulary acquisition. This chapter clarifies how understanding multilevel IP can improve our understanding of IP across the board. Section 1 defines multilevel IP and clarifies its overarching role when learners are exposed to input. Section 2 summarizes research on sentence-level IP and identifies areas in which research on lexical IP can be informative. Section 3 summarizes research on lexical IP, noting areas in which research on sentence-level IP can be informative. Finally, Section 4 discusses challenges to be met in new efforts to investigate and advance theory related to multilevel IP.
Trials-to-criterion – a method commonly used in applied behavior analysis as an alternative to the pre-/post-test study design – has been used to observe and assess input processing in processing instruction (PI) research. The results from trials-to-criterion studies have made important contributions to our understanding of the effects of PI in input processing, particularly regarding the role of explicit information. In order to promote trials-to-criterion as a method in PI research, this chapter reviews and discusses studies that have used trials-to-criterion data to investigate the effects of PI on second language processing. It also examines crucial design choices in the implementation of trials-to-criterion as a measure of language processing. Finally, it discusses future considerations for trials-to-criterion research.
We examined whether French L2 learners trained with processing instruction (PI) or traditional instruction (TI) on the French causative, would transfer training effects to the passive construction. Experiment 2 included explicit information (EI) but not Experiment 1. Accuracy results and eye movement patterns showed PI had a positive impact on participants’ processing of the causative, but neither treatment had any effect on the passive. The null effect could be attributed to the low number of tokens used and to the nature of the passive structure. Furthermore, PI training on the causative might have trained learners to rely on the necessary cues to process causative sentences correctly, but this might not have been sufficient to decrease reliance on the first noun principle.
Research on processing instruction (PI) has shown that PI is an effective instructional intervention for learners of various languages and contexts. This chapter addresses a basic question: Why does PI work? In the first section, I provide an overview of the language faculty and discuss how PI differs from other research in terms of the nature of language. The second section explores how VanPatten’s model of input processing (VanPatten, 1996, 2004, 2020) can be understood within larger frameworks that integrate language processing and acquisition. The third section discusses the characteristics of PI and the findings of PI research within the wider context of growth within the language faculty. Finally, I conclude with thoughts on future directions of PI research.
Evidence for the input processing (IP) model (VanPatten, 1996, 2004, 2007, 2015a, 2020) and the benefits of processing instruction (PI) comes largely from offline tasks, despite the fact that both are concerned with online sentence comprehension. This chapter aims to stimulate more online research in these areas. I first provide a brief summary of techniques that measure real-time sentence comprehension and then offer suggestions for how they can be used in research on the IP model and PI. Lastly, I review recent online studies of the effects of PI on second language (L2) learners’ real-time sentence comprehension. I end by proposing two new hypotheses regarding the potential that PI has to benefit L2 learners’ online processing of input.