In processing filler-gap dependencies, comprehenders quickly postulate gaps in syntactically licensed positions, but not in syntactic islands. This suggests that comprehenders can accurately use syntactic constraints to guide processing. However, resumptive pronouns appear to challenge this generalization. Resumption is ungrammatical in English. Nevertheless, they appear to immediately allow resolution of a filler dependency in syntactic islands (
To interpret a sentence like (1), the filler (
(1) | Dale solved |
Many studies have demonstrated that the gap site is constructed shortly after the filler is identified, a process called active dependency formation (
(2) | My brother wanted to know |
Importantly, active gap formation appears to be suppressed in syntactic islands (
(3) | a. | The teacher asked |
b. | The teacher asked |
Occasionally, fillers associate with a resumptive pronoun instead of a gap. For instance, in (4a),
(4) | a. | St. Louis has |
b. | *St. Louis has |
Traditionally, resumptive pronouns are described as syntactic mechanisms for repairing sland violations (
In a self-paced reading study, Hofmeister & Norcliffe (
The finding that resumptive dependencies appear to be constructed on-line challenges the generalization that filler dependency processing is guided by grammatical constraints. First, it immediately raises the question of why comprehenders do not search for resumptive pronouns in syntactic islands generally. For instance, as described above, active gap formation appears to be suppressed in syntactic islands. However, if comprehenders may actively search for resumptive pronouns in syntactic islands, then it is unclear why active dependency formation processes should be island-sensitive, since resumptive pronouns commonly appear in these configurations.
This is further underscored by “McCloskey’s Generalization”, which observes that resumptive pronouns are morphologically identically to anaphoric pronouns (
(5) | Dale solved |
Previous approaches have suggested that resumption facilitates the processing of filler dependencies by helping the comprehender identify the intended resolution site in complex structures (
To resolve this tension, Chacón (
In a sentence completion task, Chacón (
(6) | The bridesmaid speculated |
He interpreted this as implying that the search for a gap was abandoned if the pronoun referred to the filler. Importantly, this effect was modulated by the availability of other potential antecedents for the pronoun. He found that the addition of a masculine referent, such as replacing the
The assumption that comprehenders are capable of abandoning an expected gap conflicts with findings on the processing of multiple-gap constructions, such Across-The-Board (ATB) configurations. Grammatical principles require that if a filler binds a gap in one conjunct, it must also bind a gap in all subsequent conjuncts (
(7) | a. | The man that Dale arrested ___ and Harry interrogated ___. |
b. | *The man that Dale arrested Ben and Harry interrogated ___. | |
c. | *The man that Dale arrested ___ and Harry interrogated Ben. |
Wagers & Phillips (
(8) |
In this paper, I elaborate on this proposal, by arguing that the conditions of resumptive dependency formation partially follow from an interaction between anaphoric processes and disrupted active gap formation processes. I submit that shortly after encountering a filler, the comprehender builds a representation of a structure that contains a gap, e.g., a VP dominating a trace. This representation must be maintained in working memory, until the filler and the gap are associated. Thus, in my view, active gap formation depends in part in maintaining an active representation of upcoming structure that hosts the gap site, and incrementally determining whether this prediction has been satisfied.
In many models, representations that are stored in memory are vulnerable to being lost. Shunting new information into the focus of attention may displace representations that are stored in working memory, or may otherwise lead to interference effects that result in difficulty of maintaining a information in working memory (
Sketch of the proposal. If the comprehender maintains a prediction of upcoming gapped structure, then mismatch between bottom-up input and predicted structure is a cue that the sentence is ill-formed. If the predicted structure degrades in memory, then the ungrammaticality goes undetected. This allows coreference between the filler and pronoun, which enables a coherent semantic interpretation.
The first major prediction is that increasing demand on working memory overall should reduce sensitivity to the ungrammaticality of an unresolved filler dependency. Upon encountering a filler, the comprehender generates a predicted structure containing a gap, which is stored in memory. If the bottom-up input mismatches this prediction, then the sentence is perceived to be unacceptable. However, if this representation sufficiently degrades, then the comprehender will be unable to detect that the sentence is ill-formed.
This component of my proposal shares similarities with “forgetting effects” in the processing of multiple center-embedded structures (
The second major prediction is that, in contexts in which the predicted gap has decayed, a “pronoun-driven” coreference relation between the filler NP and pronoun aids coherence. When the predicted gapped structure decays, there is no syntactically licensed mechanism for relating the filler NP to the meaning of the sentence. However, if the comprehender selects the filler NP as the pronoun’s antecedent, then the interpretation of the sentence is recovered. Thus, acceptability should increase when coreference between the filler NP and the pronoun is favored, but only when sensitivity to the syntactic ill-formedness of the dependency had already diminished. I argue that these two factors independently conspire to increase acceptability of a filler associating with a pronoun in syntactically complex configurations, which occasionally results in the sentence being perceived as acceptable.
To clarify what I mean by coherence, examine (9a). In this sentence, the filler does not bind a gap, resulting in an unresolved filler dependency. Moreover, in (9a), the filler
(9) | a. | |
The maid said that this is the butler that her friend really likes kids. | ||
b. | ||
The butler said that this is the babysitter that her friend really likes kids. |
I share with previous approaches the claim that processing difficulty is a crucial factor to the acceptability of resumption in English. However, I depart from previous approaches in two crucial ways. First, I do not characterize resumption as a mechanism for salvaging a dependency in syntactic islands/complex structures. Instead, difficult-to-process structures interfere with the memory representation of the predicted resolution site for the filler. This causes the comprehender to be less sensitive to unresolved dependencies generally. Put differently, resumption does not aid difficult-to-process structures. Instead, difficult-to-process structures enable resumption, in part by disrupting the comprehender’s ability to discharge the filler dependency in a syntactically licensed way.
Secondly, this proposal makes no reference to islandhood as such. The significant majority of work on resumption in theoretical syntax (
Finally, a note on cross-language variation. This paper focuses on English, in which resumption is generally ungrammatical. This contrasts with languages like Irish, Hebrew, and Arabic, in which resumption is grammatically licensed (
In this paper, I test three predictions of this analysis. First, I examine the influence that length has on acceptability of resumption. Longer dependencies should favor resumption, as previously shown (
I present on the results of four judgment studies. Experiments 1 and 2 were speeded acceptability judgment tasks, and Experiments 3 and 4 used a new experimental paradigm. In these experiments, participants judged sentences while simultaneously maintaining a list of words in memory. In the critical stimuli, filler dependencies did not resolve with a predicate containing a gap (a
In Experiment 1, I show that participants robustly prefer gapped resolutions for filler dependencies. In Experiment 2, I show that increased length between the filler and the embedded predicate favors resumption. In Experiment 3 and 4, I show that the addition of a working memory task overall raises the acceptability of gapless resolutions, although the relative patterns in Experiments 1 and 2 are replicated in Experiments 3 and 4. In a meta-analysis of these four studies, I then suggest that the pattern of results within and between experiments are consistent with the proposal here, and show that memory strain crucially diminishes sensitivity to gapless resolutions, which may lead to a preference for resumption.
There were two goals for Experiment 1. The first goal was to demonstrate that gapped predicates are strongly preferred to gapless ones if a sentence contains a filler dependency. The second goal was to determine whether the availability of resumption would improve ratings for gapless predicates. In Experiment 1, I purposefully constructed materials that contained short filler dependencies, which should be unlikely to decay in memory. Thus, I predicted that participants would overwhelmingly reject fillers with gapless predicates, i.e., there should be no detectable effect of resumption in Experiment 1. In Experiments 2–4, I changed aspects of this design, which resulted in increased acceptance of resumption.
Previous work on resumption in English (
For instance, in an oral judgment task, McKee & McDaniel (
In contrast to these findings, Alexopoulou & Keller (
In contrast to these findings, Han et al. (
Finally, Beltrama & Xiang (
Thus, although the pattern of acceptance rates for resumption is mixed, resumption appears to generally be disfavored in English. In all these cases, the crucial comparisons were between a gap and a resumptive pronoun in the same syntactic position. From my perspective, the more crucial question is whether the comprehender prefers resolving a filler dependency with a pronoun over the possibility of a later, well-formed gap, and how acceptability of resumption compares to the baseline of a filler dependency being unresolved entirely. For this reason, the crucial manipulation in Experiments 1–4 was the presence of a gap after a (potentially) resumptive pronoun, and the possible interpretations of this pronoun. If resumption in English is ungrammatical, then it was predicted that gapped sentences would be overwhelmingly preferred to gapless sentences, regardless of whether a resumptive pronoun was present. However, if resumption is grammatical, then I predicted that comprehenders may be more likely to accept the pronoun as a suitable resolution site for the dependency if the pronoun and filler NP corefer.
There were 53 participants recruited for Experiment 1 from Amazon’s Mechanical Turk platform (
For Experiment 1, I designed 36 sets of target stimuli, and 32 filler items. The fillers were 50% grammatical. Most filler sentences contained two clauses, and most contained some kind of long-distance dependency, such as a cleft, pseudo-cleft, or relativization. Ungrammatical sentences had a variety of errors, but mostly used subcategorization errors, or local morphosyntactic errors such as agreement or case mismatches. Some filler items had multiple errors.
For the target stimuli, I manipulated ±Gap and Pronoun Reference (Ambiguous, Filler, Subject), yielding a 2 × 3 design. All target sentences contained a cleft (focused relativization) dependency and a pronoun. The gender of the pronoun was counterbalanced across all conditions across items. The pronoun was always a possessor in the subject of the embedded clause. This position was chosen because extraction from genitives is not allowed in English, and English-speakers readily produce resumptive pronouns in genitive NP positions (
For the ±Gap factor, I manipulated whether the end of the sentence contained a gap for the open filler dependency (+Gap) or not (–Gap). In the +Gap conditions, I used verbs that I judged to have a strong bias for taking an NP object, and that could plausibly take the filler as an argument. For the –Gap conditions, I used verbs that had a strong bias against taking an NP object, or transitive constructions that had NP objects. This was to ensure that comprehenders could quickly detect that the filler was not an argument of the critical verb upon entering the VP, at least if they still maintained a prediction for a gapped structure. All target materials were matched in length.
Each target stimulus contained two NPs before the critical pronoun: the subject NP of the main clause, and the filler NP. Both NPs could be potential antecedents for the critical pronoun. I manipulated the stereotypical genders of these NPs, using the strongly masculine and feminine stereotyped nouns from the norming study in Kennison & Trofe (
(10) | a. | |
The maid[F] said that this is the babysitter[F] that |
||
b. | ||
The butler[M] said that this is the babysitter[F] that |
||
c. | ||
The maid[F] said that this is the butler[M] that |
||
d. | ||
The maid[F] said that this is the babysitter[F] that |
||
e. | ||
The butler[M] said that this is the babysitter[F] that |
||
f. | ||
The maid[F] said that this is the butler[M] that |
I predicted that there should be an effect of ±Gap, with +Gap conditions strongly preferred to –Gap conditions. This is because I understood resumption to be ungrammatical in English, and thus the addition of a pronoun would likely not reduce the “penalty” for gapless sentences. However, if comprehenders could resolve the filler dependency with the pronoun, then acceptance rates should increased for the –Gap, Ambiguous and –Gap, Filler sentences compared to the –Gap, Subject sentences. If resumptive dependencies are only constructed in contexts of high memory demand, then I did not expect the pronoun’s interpretation to have much effect in Experiment 1. This is because the sentences were specifically designed to not include structures that introduced exceptional processing difficulty, even though they were contained in a syntactic island. In other words, I predicted no interaction effects between ±Gap and Pronoun, given that processing demands were designed to be minimal in Experiment 1.
For Experiment 1, I used a speeded acceptability judgment task. I chose this task because it has been used to reveal “grammatical illusions”, or sentences that are ungrammatical but are momentarily perceived to be acceptable (e.g.,
Participants were recruited from Amazon Mechanical Turk. Upon accepting the HIT (Human Interaction Task) on the Mechanical Turk platform, participants were instructed to navigate to the experiment hosted on the IbexFarm (
Sentences were displayed word-by-word centered on the screen, using a rapid serial visual presentation (RSVP) design. Each word was displayed for 300 ms. Each sentence was preceded with a fixation cross and followed by a period, both of which were displayed for 500 ms. After the sentence ended, participants were asked whether the sentence was acceptable. They entered their answer either by clicking on a “Yes” or “No” button displayed on the screen, or by using the 1 and 2 keys. All experimental materials were presented in a randomized order, distributed in six separate lists in a 2 × 3 design.
After completing the task, participants were given a code that they entered into Mechanical Turk. They were compensated $2.00 for participation, and it took approximately 20 minutes to complete the task.
For the filler items, I constructed a logit mixed-effects model using Grammaticality as the factor, and random effects for participant and item.
For the target items, I constructed a logit mixed-effects models using the
Results of logit mixed-effects model fit to the acceptability data for the target items in Experiment 1. Starred (*) rows correspond to
SE | |||||
---|---|---|---|---|---|
Intercept | –0.18 | 0.21 | –0.86 | 0.39 | |
Gap | 2.69 | 0.23 | 11.73 | <0.01 | * |
Pronoun:1 | 0.13 | 0.21 | 0.62 | 0.54 | |
Pronoun:2 | –0.01 | 0.20 | –0.06 | 0.95 | |
Gap × Pronoun:1 | 0.25 | 0.21 | 1.23 | 0.22 | |
Gap × Pronoun:2 | –0.28 | 0.20 | –1.41 | 0.16 |
Results of pairwise comparisons for the model fit to the acceptability judgments of Experiment 1. Comparisons were made between the three levels for the Pronoun factor nested within each level of the ±Gap factor.
SE | |||||
---|---|---|---|---|---|
+Gap | Ambiguous – Filler | 0.68 | 0.47 | 1.43 | 0.32 |
Ambiguous – Subject | 0.47 | 0.47 | 1.01 | 0.57 | |
Filler – Subject | –0.20 | 0.46 | –0.44 | 0.90 | |
–Gap | Ambiguous – Filler | –0.39 | 0.51 | –0.76 | 0.73 |
Ambiguous – Subject | 0.02 | 0.57 | 0.03 | 1.00 | |
Filler – Subject | 0.41 | 0.55 | 0.75 | 0.74 |
Mean acceptance rates by condition from Experiment 1, with error bars representing one standard error from the mean.
Most strikingly, the +Gap conditions were overwhelmingly accepted, and the –Gap conditions were overwhelmingly rejected. This was predicted, because I assumed that resumption is ungrammatical in English, i.e., participants would only accept sentences in which the filler dependency resolves with a gap. Participants did not appear to treat the pronoun in the subject NP as a suitable resolution site for the cleft dependency.
Performance on Experiment 1 could also be understood as a signal detection task, i.e., correctly distinguishing grammatical sentences from ungrammatical sentences. For this reason, I computed the sensitivity index, or d′, to assess accuracy. D′-scores are ‘a measure of a participant’s ability to discriminate between the signal, while taking into account participant response biases (
The goal of Experiment 1 was to determine whether the availability of resumption improved the acceptability of gapless sentences. The results from Experiment 1 found no such improvement. Previous work demonstrated that sentences with resumptive dependencies often were assigned low ratings in off-line judgments (
One potentially surprising finding was that there was no detectable weak cross-over effect in Experiment 1 (
(11) | *Who |
In a real-time processing study, Kush et al. (
In summary, the results of Experiment 1 failed to demonstrate any facilitatory effect for resumption, even under time pressure. However, this is unsurprising. I purposefully designed these materials to have shorter filler dependencies, and I purposefully selected a syntactic island that I supposed would not induce significant strain on the participants’ memory. In Experiment 1, the filler NP and the pronoun were only separated by one word, the complementizer
In Experiment 2, the position of the cleft filler NP and the non-filler subject NP were changed, such that the cleft was in the main clause and the non-filler subject NP was in the embedded clause. This increased the length of the cleft dependency, and also placed the filler NP as a prominent argument in the main clause, making it a more accessible antecedent for the pronoun. These changes were predicted to make the resumptive analysis more likely.
As in Experiment 1, the goal of Experiment 2 was to determine whether the availability of resumption improved the acceptability of gapless sentences. In Experiment 1, there was no evidence that resumption improved acceptability. However, the structure of the materials in Experiment 1 discouraged resumption, due to the length of the filler dependency and the syntactic positions of the possible NP antecedents for the pronoun.
In Experiment 2, the order of the two NPs was reversed, such that the clefted filler NP occurred in the main clause, and the non-filer subject NP occurred in the embedded clause. I predicted that the reverse order would make the filler NP more syntactically prominent. This is important, because pronouns more easily select syntactically prominent antecedents (
There were 60 participants recruited for Experiment 2 from Amazon’s Mechanical Turk platform, using the same inclusion criteria as in Experiment 1. All participants self-identified as native English-speakers, and the mean age of participants was 36.
The materials in Experiment 2 were similar to the those in Experiment 1, except that the filler NP and the non-filler subject NP were re-ordered, such that the filler NP appeared first in the main clause, and the non-filler subject NP appeared second in the embedded clause. This change made the filler NP more syntactically prominent, and increased the length of the cleft dependency. The materials for Experiment 2 are exemplified in (12).
(12) | a. | |
This is the babysitter[F] that the maid[F] said that |
||
b. | ||
This is the babysitter[F] that the butler[M] said that |
||
c. | ||
This is the butler[M] that the maid[F] said that |
||
d. | ||
This is the babysitter[F] that the maid[F] said that |
||
e. | ||
This is the babysitter[F] that the butler[M] said that |
||
f. | ||
This is the butler[M] that the maid[F] said that |
The methods in Experiment 2 were the same as in Experiment 1. Participants were compensated $2.00 for participation, and took approximately 20 minutes to complete the task.
For the filler items, I constructed a logit mixed-effects model using Grammaticality as the factor, and random effects for subject and item, as in Experiment 1. Grammaticality was significant (
Due to an experimenter error, 40 observations out of 2160 target items had to be discarded. After excluding these observations, I constructed a logit mixed-effects model using the same procedure as in Experiment 1. The mean acceptance rates by condition are given in Figure
Results of logit mixed-effects model fit to the acceptability judgments for the target items in Experiment 2. Starred (*) rows correspond to
SE | |||||
---|---|---|---|---|---|
Intercept | 0.66 | 0.18 | 3.62 | <0.01 | * |
Gap | 1.09 | 0.13 | 8.46 | <0.01 | * |
Pronoun:1 | 0.13 | 0.11 | 1.23 | 0.22 | |
Pronoun:2 | 0.08 | 0.10 | 0.09 | 0.93 | |
Gap × Pronoun:1 | 0.18 | 0.11 | 1.68 | 0.10 | |
Gap × Pronoun:2 | –0.33 | 0.10 | –3.23 | <0.01 | * |
Results of pairwise comparisons for the model fit to the acceptability judgments from Experiment 2. Comparisons were made between the three levels for the Reference factor nested within each level of the ± Gap factor. Starred (*) rows correspond to
SE | ||||||
---|---|---|---|---|---|---|
+Gap | Ambiguous – Filler | 0.65 | 0.28 | 2.31 | 0.06 | |
Ambiguous – Subject | 0.28 | 0.30 | 0.92 | 0.63 | ||
Filler – Subject | –0.37 | 0.26 | –1.42 | 0.33 | ||
–Gap | Ambiguous – Filler | –0.37 | 0.21 | –1.7 | 0.18 | |
Ambiguous – Subject | 0.23 | 0.23 | 1.00 | 0.58 | ||
Filler – Subject | 0.60 | 0.25 | 2.44 | 0.04 | * |
Mean acceptance rates by condition from Experiment 2, with error bars representing one standard error from the mean. Reported
As in Experiment 1, the most striking result is that the +Gap sentences were accepted significantly more often than the –Gap sentences. However, unlike in Experiment 1, there was a significant interaction between ±Gap and Pronoun, which was reflected in the increased acceptance rates for the –Gap, Filler condition compared to the –Gap, Subject condition. Thus, in Experiment 2, there appeared be a facilitation effect of resumption, which was not observed in Experiment 1.
The average d′-score across participants and all items for Experiment 2 overall was 1.61 ± 0.11. For filler items, the average d′-score was 2.27 ± 0.14. For target items, the mean score was 1.40 ± 0.15. This suggests that participants were moderately capable in discriminating between grammatical and ungrammatical items.
The goal of Experiment 2 was to determine whether resumption would improve sentences with longer cleft dependencies than the sentences used in Experiment 1. In Experiment 1, there was no evidence that reference between the filler NP and pronoun improved acceptability, suggesting that comprehenders did not entertain a resumptive analysis. In Experiment 2, coreference between the filler NP and pronoun was facilitated, because the filler NP was in a syntactically more prominent position, which increased the likelihood of coreference between the pronoun and the filler (
In Experiments 3 and 4, I added the word list recall task to the speeded acceptability judgment task. This was designed to strain working memory resources while participants processed and judged a sentence. On the analysis that I propose, increased strain on memory sources was predicted to increase the acceptability of –Gap target items overall. This is because the additional strain on memory should result in quicker decay of the prediction of a gapped structure in memory. This then should result in diminished ability to determine whether a filler dependency was resolved in a syntactically licensed way, i.e., with a gap.
The goal of Experiment 3 was to determine whether the acceptability of resumption increased with an increased strain on memory. As described earlier, many previous accounts of resumption have proposed that processing difficulty and syntactic complexity partially determine the distribution of resumption (
To test this, I introduced a word list recall task in Experiments 3 and 4. Participants were asked to memorize a list of words before judging a sentence as acceptable or unacceptable, and then respond to a probe word after judging the sentence. Crucially, the materials in Experiment 3 were identical to those in Experiment 1. On my proposal, the addition of this task should lead to a degraded representation of the gapped structure in memory, resulting in less sensitivity to the ungrammaticality of an unresolved filler-gap dependency.
There were 60 participants recruited for Experiment 3 from Amazon’s Mechanical Turk platform (
The sentences that participants were asked to judge were the same as in Experiment 1. For the word list recall task, there were three short nouns not mentioned in the stimulus sentence that preceded it. These nouns were randomly selected from a dictionary, and then hand-selected to avoid repetition between the word-list and the target sentence. I also avoiding using nouns that had clear semantic relations to the sentence or to each other. Most nouns were monomorphemic.
The methods for Experiment 2 were the same as in Experiment 1, with the exception of the word list recall task. Before each trial, participants saw the list of words displayed in the center of the screen for 1000 ms. Afterwards, the sentence to be judged was automatically displayed in an RSVP design as in Experiments 1 and 2. Then, participants were asked to judge the sentence as acceptable or unacceptable. After judging it, participants were asked whether a probe word was in the initial list of words that they were asked to memorize. On half of the trials, the probe word was in the word list before the judgment phase. Participants were given as much time as necessary to provide a judgment for the sentence and for responding to the probe word. This process is illustrated in Figure
Method for Experiment 3. First, the word list was displayed. Then, the speeded acceptability judgment task was conducted. Afterwards, participants were asked whether a probe word was contained in the initial list.
In this section, I first describe performance on the word list recall task, and then performance on the acceptability judgment task. For the acceptability results, I first conducted analysis on the raw results. Afterwards, I reanalyzed the dataset, excluding trials with incorrect responses on the word list recall task. I decided to do this, because participants may have primarily attended to one of the two tasks. Primarily attending to the judgment task may produce results more similar to those in Experiment 1, but with low accuracy rates on the word list recall task. Conversely, attending to the word list recall task may produce noisier performance on the acceptability judgment task, or it may amplify the effect of memory strain on the acceptability judgments. Because I am interested in the effect of memory strain on judgments in Experiment 3, I chose to err on the side of excluding trials in which participants were not attending to the word list recall task. For transparency, I report on both results, and highlight any crucial differences.
For probe word recall accuracy, I report on both percentage of correct trials and average d′-scores across participants. Overall, probe word recall accuracy was 80.4 ± 0.4%. The average d′-score across all items and participants was 1.92 ± 0.10, suggesting moderate accuracy in the recall task in Experiment 3. For the filler items, probe word recall accuracy was 80.0 ± 0.06%. For Grammatical filler items, recall accuracy was 81.1 ± 0.9%, and for Ungrammatical filler items, recall accuracy was 78.9 ± 0.9%. In d′-scores, recall accuracy for filler items was 2.2 ± 0.17. For target items, the mean accuracy was 80.8 ± 0.06%. Mean accuracy by condition for target items is broken down in Table
Mean accuracy and standard error on word list recall task by condition for Experiment 3.
+Gap | –Gap | |
---|---|---|
78.1 ± 1.5% | 76.9 ± 1.6% | |
81.1 ± 1.5% | 78.9 ± 1.5% | |
85.6 ± 1.3% | 83.9 ± 1.4% |
Next, I describe the acceptance rates in the acceptability judgment task. First, as before, I constructed a logit mixed effects model to analyze the filler items, with the same structure as Experiments 1 and 2. Grammatical filler items were accepted more often than Ungrammatical filler items, in both the dataset including incorrect trials on the word list recall task (
For the target items, I constructed a logit mixed-effects model for both datasets with the same structure as in Experiments 1 and 2, and conducted the same pairwise comparisons. The mean acceptance rates by condition after exclusion are plotted in Figure
Mean acceptance rates by condition in Experiment 3, with error bars representing one standard error from the mean. Reported
Results of logit mixed-effects model fit to the acceptability judgments for the target items with correct word list recall in Experiment 3. Starred (*) rows correspond to
SE | |||||
---|---|---|---|---|---|
Intercept | 0.95 | 0.34 | 2.81 | <0.01 | |
Gap | 2.90 | 0.33 | 8.87 | <0.01 | * |
Pronoun:1 | 0.24 | 0.33 | 0.74 | 0.46 | |
Pronoun:2 | –0.75 | 0.26 | –2.90 | <0.01 | * |
Gap × Pronoun:1 | –0.28 | 0.28 | –1.02 | 0.31 | |
Gap × Pronoun:2 | –0.44 | 0.26 | –1.69 | 0.09 |
Results of pairwise comparisons for the model fit to the acceptability judgments for target items with correct word list recall in Experiment 3. Comparisons were made between the three levels for the Reference factor nested within each level of the ± Gap factor. Starred (*) rows correspond to
SE | ||||||
---|---|---|---|---|---|---|
+Gap | Ambiguous – Filler | 1.15 | 0.68 | 1.70 | 0.21 | |
Ambiguous – Subject | –1.27 | 1.12 | –1.14 | 0.49 | ||
Filler – Subject | –2.42 | 0.96 | –2.53 | 0.04 | * | |
–Gap | Ambiguous – Filler | 0.84 | 0.53 | 1.57 | 0.26 | |
Ambiguous – Subject | 0.74 | 0.53 | 1.40 | 0.34 | ||
Filler – Subject | –0.11 | 0.56 | –0.19 | 0.98 |
As in Experiments 1 and 2, the most striking fact is the main effect of ±Gap, with the +Gap conditions accepted significantly more often than the –Gap conditions. Additionally, there was a main effect of Pronoun:2. In the dataset with incorrect memory trials excluded, pairwise comparisons revealed a dispreference for coreference between a filler NP and the pronoun within the +Gap level. This is consistent with the weak cross-over effect described in Section 2.6, since it reflects a bias against a filler NP binding both a pronoun and a gap in the same sentence. However, no comparisons were significant within the level –Gap.
The model fit to the dataset that included incorrect trials on the word list recall task showed a similar profile, with a main effect of ±Gap (
Before exclusion, the average d′ score across all participants and items was 1.81 ± 0.14, and after exclusion, 1.84 ± 0.13. Thus, participants were moderately successful at discriminating between grammatical and ungrammatical sentences. For the filler items, the average d′-scores across participants was 1.86 ± 0.14 before exclusion, and 1.85 ± 0.13 after exclusion. For the target items, the average d′-score was 2.46 ± 0.26 before exclusion, and 2.62 ± 0.25 after exclusion.
Participants performed accurately on the word list recall task, while maintaining sensitivity to the distinction between grammatical and ungrammatical sentences, both in the filler items and in the target items. This demonstrated that participants were capable of maintaining the word list in memory while simultaneously processing the complex sentences. Interestingly, participants were more accurate at the word list recall task in the Subject conditions than other target items, which was unpredicted.
In Experiment 1, participants overwhelmingly rejected the –Gap sentences, regardless of the pronoun’s interpretation. This was likely because the materials were designed to not induce significant demand on memory resources. By contrast, it was predicted that the addition of the word list recall task would result in increased acceptability for the –Gap sentences and for resumption generally. However, there was no specific increase in judgments for the –Gap, Subject sentences in the analysis of either dataset, as I had predicted. One possibility is that dependency length is a precondition for resumption in English. Thus, although participants were placed under memory strain in Experiment 3, they did not analyze the pronoun as resumptive, due to the dependency being too short. Relatedly, length may be a precondition for resumption because decay is a significant factor to degradation of the representation in memory. In other words, the addition of the word list recall task was not sufficiently difficult enough to engender resumption. Alternatively, participants may not have considered a resumptive analysis due to relative salience of the filler NP as an antecedent. Like Experiment 1, the filler NP was not located in the main clause. Thus, the non-filler subject NP may been significantly more accessible, and therefore participants were less able to access the filler NP as an antecedent.
The account of resumption sketched in this paper also predicted that –Gap sentences may be accepted more frequently overall with the addition of the word list recall task. Qualitatively, the –Gap target item acceptance rates increased compared to Experiment 1. However, as in Experiment 2, the acceptance rates for Ungrammatical filler sentences also increased, suggesting that participants may have simply guessed more frequently in Experiment 3. I turn to this issue in Section 6.
Next, the goal of Experiment 4 was to further explore the interaction between memory strain and length. In Experiment 4, I used the same items as in Experiment 2. These items were designed to favor resumption. If the ambiguity advantage in Experiment 3 resulted from failure to maintain both possible referents, then it is predicted that a similar ambiguity advantage should surface in Experiment 4.
As in Experiment 3, the goal of Experiment 4 was to determine whether the additional memory strain would result in higher rate of acceptance of resumptive dependencies, as reflected in higher ratings for –Gap, Subject sentences compared to –Gap, Subject sentences. In Experiment 4, I used the same materials from Experiment 2, which had proven more likely to engender a resumptive interpretation.
There were 60 participants recruited for Experiment 4 from Amazon’s Mechanical Turk platform (
The sentences that participants were asked to judge were the same as in Experiment 2. For the word list recall task, the same three short inanimate nouns that were generated in Experiment 3 were used.
The methods for Experiment 4 were the same as in Experiment 3. Participants were compensated $
As in Section 4.5, I first report on the word list recall results, and then the acceptability judgment results. I first conducted analysis on the raw acceptability results, and then excluding trials with incorrect memory recall.
Overall, the mean word list recall accuracy was 80.6 ± 0.4%. The average d′-score across all items and participants was 2.06 ± 0.14. This suggests that there was moderately high accuracy in the recall task. For the filler items, recall accuracy was 81.0 ± 0.7% overall. For Grammatical filler items, the recall accuracy was 82.0 ± 0.9%, and for Ungrammatical filler items, the recall accuracy was 80.0 ± 0.9%. In d′-scores, memory recall accuracy for filler items was 2.28 ± 0.17. For target items, the mean accuracy was 80.3 ± 0.6%. The mean accuracy by condition for target items is broken down in Table
Mean accuracy and standard error on word list recall task by condition for Experiment 4.
+Gap | –Gap | |
---|---|---|
78.2 ± 0.02% | 78.6 ± 0.02% | |
81.3 ± 0.02% | 79.6 ± 0.02% | |
85.0 ± 0.01% | 85.0 ± 0.01% |
The acceptability judgment data was analyzed with the same methods as in Experiments 1–3. For the filler items, Grammatical filler items were accepted more often than Ungrammatical filler items, in both the dataset including incorrect trials on the memory recall task (
For the target items, I constructed a logit mixed-effects models for both datasets with the same structure as in Experiments 1–3. The mean acceptance rates by condition after exclusion are plotted in Figure
Mean acceptance rates by condition in Experiment 4, with error bars representing one standard error from the mean. Reported
Results of logit mixed-effects model fit to the acceptability judgments for target items in Experiment 4. Starred (*) rows correspond to
SE | |||||
---|---|---|---|---|---|
Intercept | 1.79 | 0.29 | 6.10 | <0.01 | * |
Gap | 1.56 | 0.27 | 5.69 | <0.01 | * |
Pronoun:1 | 0.11 | 0.28 | 0.39 | 0.70 | |
Pronoun:2 | 0.02 | 0.25 | 0.07 | 0.95 | |
Gap × Pronoun:1 | 0.20 | 0.26 | 0.77 | 0.44 | |
Gap × Pronoun:2 | –0.54 | 0.23 | –2.33 | 0.02 | * |
Results of pairwise comparisons for the model fit to acceptability data from the target items with correct word list recall from Experiment 4. Comparisons were made between the three levels for the Reference factor nested within each level of the ± Gap factor. Starred (*) rows correspond to
SE | ||||||
---|---|---|---|---|---|---|
+Gap | Ambiguous – Filler | 0.84 | 0.80 | 1.05 | 0.55 | |
Ambiguous – Subject | 0.10 | 0.84 | 0.12 | 0.99 | ||
Filler – Subject | –0.74 | 0.74 | –0.99 | 0.58 | ||
–Gap | Ambiguous – Filler | –0.65 | 0.39 | –1.62 | 0.21 | |
Ambiguous – Subject | 0.37 | 0.45 | 0.82 | 0.69 | ||
Filler – Subject | 1.02 | 0.40 | 2.56 | 0.03 |
As in the previous three experiments, there was a strong preference for +Gap target items over –Gap target items. Unlike in Experiment 3 however, there was no main effect of Pronoun. There was a significant interaction between ±Gap and Pronoun, which was reflected in the increased acceptance rates for the –Gap, Filler sentences compared to the –Gap, Subject sentences. Thus, in Experiment 4, resumptive pronouns detectably facilitated the acceptability of a sentence, as in Experiment 2. Unlike Experiment 3, there was no evidence of the weak cross-over effect.
The model fit to the data that did not exclude incorrect memory recall trials had a similar pattern. There was a main effect of ±Gap (
The average d′ across all items and participants was 1.55 ± 0.14 before exclusion incorrect recall trials, and 1.96 ± 0.16 after exclusion. The average d′-score for filler items was 2.25 ± 0.20 before exclusion, and 2.36 ± 0.22 after exclusion. The d′-scores for all target items was 1.34 ± 0.20 before exclusion, and 1.31 ± 0.22 after exclusion.
The goal of Experiment 4 was to determine whether acceptance rates increased for sentences with a possible resumptive analysis. The general pattern of results from Experiment 2 were replicated in Experiment 4, i.e., –Gap, Filler target items were accepted more often than –Gap, Subject items. This is unsurprising, since the same materials were used across the two experiments.
Qualitatively, the results of Experiment 4 suggested a greater acceptance rates for the –Gap target items compared to Experiments 1–3. On the proposal in this paper, the length of the filler dependency and the word list recall task both strain memory resources, which results in a diminished ability to detect the ungrammatical resolution of the filler dependency. I turn to the cross-experiment comparisons in the next section to demonstrate that this is a reliable pattern across Experiments 1–4.
The proposal in this paper makes several predictions. The first prediction is that increased length between a filler NP and a pronoun should favor resumption, which has been demonstrated in previous studies (
In this section, I demonstrate that the cross-experiment comparisons are quantitatively reliable. Moreover, I suggest that these contrasts cannot be due to overall noisier performance, corresponding to increased task difficulty. First, I conduct a meta-analysis across the acceptability ratings across the four experiments. Then, I conduct a meta-analysis of the sensitivity (d′) scores. In both cases, I show that length has a significant impact on the acceptability of resumption, and that memory load and length furthermore facilitate the acceptability of ungrammatical sentences overall.
First, I examined the acceptance rates of the target items across Experiments 1–4. I collated the data from the four experiments, including the trials with incorrect word list recall performance in Experiment 3 and Experiment 4. I then constructed a logit mixed effects model with the acceptance rates of the target items as the dependent variable, and with ±Gap and Pronoun as within-participant factors. I also included two new between-participant factors, Memory Load and Length. Memory Load had levels –Memory Load (Experiments 1 and 2) and +Memory Load (Experiments 3 and 4). Length had levels Short (Experiments 1 and 3) and Long (Experiments 2 and 4). I also included random slopes for ±Gap × Pronoun by participant, and random effects by item.
Results of logit mixed-effects model fit to the acceptability judgment data for all target items in Experiments 1-4. Starred (*) rows correspond to
SE | |||||
---|---|---|---|---|---|
Intercept | 0.39 | 0.15 | 2.53 | 0.01 | * |
Gap | 1.05 | 0.13 | 7.85 | <0.01 | * |
Pronoun:1 | 0.08 | 0.10 | 0.83 | 0.40 | |
Pronoun:2 | 0.02 | 0.09 | 0.20 | 0.84 | |
Memory Load | 0.98 | 0.17 | 5.63 | <0.01 | * |
Length | –0.53 | 0.22 | –2.41 | <0.02 | * |
Gap × Pronoun:1 | 0.13 | 0.10 | 1.24 | 0.22 | |
Gap × Pronoun:2 | –0.34 | 0.10 | –3.47 | <0.01 | * |
Gap × Memory Load | –0.21 | 0.17 | –1.23 | 0.22 | |
Pronoun:1 × Memory Load | 0.17 | 0.17 | 0.98 | 0.33 | |
Pronoun:2 × Memory Load | –0.12 | 0.16 | –0.74 | 0.46 | |
Gap × Length | 1.81 | 0.21 | 8.43 | <0.01 | * |
Pronoun:1 × Length | 0.17 | 0.17 | 0.98 | 0.33 | |
Pronoun:2 × Length | –0.11 | 0.16 | –0.74 | 0.46 | |
Memory Load × Length | –0.01 | 0.27 | –0.02 | 0.98 | |
Gap × Pronoun:1 × Memory Load | 0.02 | 0.13 | 0.13 | 0.90 | |
Gap × Pronoun:2 × Memory Load | –0.01 | 0.13 | –0.01 | 0.93 | |
Gap × Pronoun:1 × Length | 0.12 | 0.17 | 0.72 | 0.47 | |
Gap × Pronoun:2 × Length | 0.02 | 0.16 | 0.14 | 0.89 | |
Gap × Memory Load × Length | –0.71 | 0.26 | –2.69 | <0.01 | * |
Pronoun:1 × Memory Load × Length | 0.10 | 0.21 | 0.45 | 0.65 | |
Pronoun:2 × Memory Load × Length | –0.11 | 0.19 | –0.55 | 0.58 | |
Gap × Pronoun:1 × Memory Load × Length | –0.24 | 0.21 | –1.17 | 0.24 | |
Gap × Pronoun:2 × Memory Load × Length | 0.10 | 0.20 | 0.52 | 0.61 |
Unsurprisingly, this meta-analysis revealed main effects of ±Gap, showing a strong preference for +Gap over –Gap. Importantly, there were also main effects of Memory Load and Length. The main effect of Memory Load signified that the addition of the word list recall task increased acceptance rates overall. Similarly, increasing the length of the filler-dependency had an overall effect of increasing acceptance rates.
There were three significant interactions. There was an interaction between ±Gap and Pronoun, one between ±Gap and Length, and a three-way interaction between ±Gap, Memory Load, and Length.
First, I conducted pairwise comparisons of the three levels in Pronoun nested within the two levels of ±Gap. This revealed the expected pattern. Within the +Gap conditions, Ambiguous items were accepted more than Filler items (
Next, I discuss the interaction effect between Length and ±Gap. Pairwise comparisons revealed that Long sentences were accepted more often than Short sentences within the level –Gap (
Next, I examined the three-way interaction between ±Gap, Length, and Memory Load in several pairwise comparisons. First, I examine the effect of Length within Memory Load and ±Gap. Within the level –Gap. Within this level, Short, +Memory Load items were assigned higher ratings than Short, –Memory Load (Experiment 1 vs. Experiment 3;
Next, I compared the effect of Memory Load nested within the two levels of Length and ±Gap. Within the level +Gap, Memory Load facilitated acceptance rates for Long sentences (
Effect of word list recall task on acceptance rates by condition. The y-axis corresponds to the difference between mean acceptance rates between Experiments 3 and 1 (top) and Experiments 4 and 2 (bottom).
One possible concern about the comparison between Experiments 1–4 is that the increased acceptance rates of –Gap target items may not reflect anything specific about the processing of filler dependencies. Rather, these results may reflect noisier performance induced by increased task difficulty. Increased guessing on acceptability trials should raise the acceptance rates of –Gap items, due to higher proportion of at-chance performance. The finding that +Gap, Long items were accepted less often than +Gap, Short items may also be consistent with this explanation, i.e., this may show that additional processing cost associated with longer filler dependencies may result in both –Gap and +Gap acceptance rates drawing closer to chance.
To distinguish these hypotheses, I next examine the distribution of d′-scores across participants in the four experiments. As I described in Section 2.6, the d′-score is a way of quantifying individual participants’ sensitivity that takes into account individual response bias. If the cross-experiment results were due to increased task difficulty resulting in noisier performance, then there should be a decline in d′-scores with the addition of the word list recall task and the longer filler dependencies. First, I discuss the d′-scores for target items. Then, I turn my attention to the d′-scores for filler items.
To test this, I constructed a linear mixed effects model with participants’ d′-scores computed over target items as the dependent variable. I included Memory Load, Length and their interaction terms as between-participant factors, and included random effects by participant.
However, this result does not distinguish between these two explanations. Increased guessing should lead to a decrease in sensitivity as the tasks become more complex across both target items and filler items. On my analysis, increased acceptance rates of –Gap target items increased the rate of false hits, which lowered d′-scores. Thus, on my analysis, it was predicted that filler items should show less reduction in sensitivity compared to the target items with the inclusion of the memory recall task, which increased task difficulty overall, and the length manipulation, which increased task difficulty for the target items only.
To test this, I fit a linear mixed effects model to the d′-scores for the filler items with the same structure. Memory Load had no effect on the d′-scores (
Comparisons of d′-scores across Experiments 1–4. Error bars correspond to one standard error from the mean. The results from Experiments 1 and 2 were coded as -Memory Load, and Experiments 3 and 4 were coded as +Memory Load. The results from Experiments 1 and 3 were coded as Short, and the results from Experiments 2 and 4 were coded as Long.
Overall, this meta-analysis suggested that the cross-experiment manipulations of memory load and dependency length affected the acceptability of –Gap sentences. Moreover, it confirmed the general pattern for preferring resumptive dependencies over anaphoric dependencies, when no syntactically licensed resolution site is available. Secondly, I argued that comparing the sensitivity scores between target items and filler items across the four experiments revealed a more extreme profile on the target items. Although performance may have been noisier with more complex experimental paradigms, overall, the effect was magnified for the –Gap sentences in the target items. This suggests that the cross-experiment manipulation of processing difficulty specifically impacted comprehenders’ ability to detect an unresolved filler-gap dependency.
Finally, it is worth pointing out that there was no interaction between Pronoun and the critical factors that manipulated processing difficulty, Memory Load and Length. My account would likely be more strongly supported with such an interaction. On my account, it is expected that the facilitatory effect of co-reference between a pronoun and a filler phrase should be increased when the comprehender is placed under processing difficulty. However, this meta-analysis did not demonstrate this. This may partially be due in part to the lack of resumptive effect in Experiments 1 and 3, and the small magnitude of the effect overall. Thus, I leave investigating this more systematically for future research.
An important and robust finding in sentence processing is that comprehenders actively construct filler-gap dependencies (
An anonymous reviewer suggests another interpretation of these results. The target items all contained a (potential) resumptive pronoun in the subject genitive position (e.g.,
Importantly, evidence for the weak cross-over effect was observed in Experiment 3, and in the larger meta-analysis. Similarly, the interaction effect between ±Gap and Pronoun in Experiment 2, Experiment 4, and the meta-analysis suggested that participants were capable of analyzing the subject genitive pronoun as resumptive. Moreover, I assumed that participants typically attempted to find an antecedent for a pronoun in the same sentence, without any other context to support alternative interpretations (cf.
Regardless, if active dependency formation processes depend on prediction, then it is still possible that comprehenders selectively generate predicted structures containing resumption. For instance, upon detecting a filler NP, participants may not generate a prediction for a structure containing a resumptive pronoun in the subject genitive position at a short distance from the filler NP, but they may generate predictions for resumption at greater distances. Importantly, this may explain the lack of increased acceptance for resumption in Experiment 3. In Experiment 3, participants were placed under memory strain with the word list recall task. However, they did not demonstrate improved tolerance for resumption, as predicted. Future research may benefit from exploration of resumptive pronouns in contexts in which they are assigned higher ratings than in subject genitive positions. For instance, the object genitive position (
The account I provide here makes some assumptions that are not shared by other work on the processing of filler dependencies. In this paper, I argued that filler-gap dependency processing depends on active maintenance of a predicted structure. Typically, this process is characterized as active maintenance of a filler NP while searching for a resolution site (
There are two reasons why it’s important that the predicted resolution site is maintained in working memory for my account. First, I argued that acceptability of resumption in part depends on a diminished sensitivity to the grammatical requirement that a filler binds a gap, as reflected in the variable acceptance rates for –Gap target items across Experiment 1–4. On my view, detecting the ungrammaticality of an unresolved filler dependency relies on comparing the predicted structure in memory against the bottom-up input. On the traditional view, it presumably relies on a failed prospective search for a gap. Moreover, in order for resumption to facilitate acceptability, the filler NP must be a sufficiently accessible antecedent for the pronoun (
To clarify the difference, consider the contrast in (13). On my proposal, (13a) is recognized as acceptable when the comprehender is able to detect that the predicted structure matches the input, and (13b) is recognized as unacceptable when the comprehender detects that the expected gap is not available. If this prediction has sufficiently degraded, then (13a) may be perceived as a selectional violation.
(13) | a. | This is |
b. | This is |
The existing evidence on active gap formation processes is largely equivocal on what information is stored in working memory. For instance, filled gap effects are sometimes characterized as supporting an account in which the comprehender maintains a representation of the filler NP in memory while prospectively searching for a gap. However, filled gap effects could follow from incrementally checking the predicted structure maintained in memory against the input sentence. Similarly, sustained anterior negativity results may reflect the active maintenance of the structured prediction, and not the filler NP as such.
One reason to suspect that the comprehender actively maintains features of the filler NP is the profile of “D-linked”
Another tension between my proposal and previous literature is the interpretation of length effects. On my proposal, length adversely affects maintenance of predicted structure in memory, and this predicted structure is required for determining whether a filler dependency has a syntactically licensed resolution. This predicts that length may diminish filled-gap effects. Wagers & Phillips (
On my account, it is unlikely that the representation of the gap completely degrades. For instance, if the syntactic prediction had completely degraded, then the –Gap target items should have been rated near ceiling. Conversely, the +Gap target items should have been accepted at lower rates. This is because all +Gap items contained verbs or prepositions that required objects. Thus, if the prediction for a gap had completely disappeared, then these items should have been perceived as selectional violations, e.g., an obligatorily transitive verb missing a required argument NP. However, even in Experiment 4, there was still a preference for +Gap over –Gap, implying that comprehenders still distinguished between grammatical and ungrammatical resolutions. This could reflect residual representation of the predicted structure in memory, at least on some trials. Even so, my account predicts that active dependency formation should diminish over time, if not vanish. However, this does not seem to match the findings by Wagers & Phillips (
Relatedly, an anonymous reviewer points out that the findings by Wagers & Phillips (
Finally, I have remained agnostic about the nature of the degradation of representations in memory and the structure of short-term memory. Classically, theories of working memory postulate privileged memory buffers for maintaining information accessible to computations (
The results of this study are likely uninformative for arbitrating between these approaches to short term memory. For instance, the effects of length could either be cast as decay of the predicted structure in memory or as interference between the predicted structure, the linguistic material processed after the filler NP, and the word list. Similarly, the effect of the word list could be cast as interference between the word list and the material stored in working memory, or as competition for limited working memory resources.
However, if the impact of the word list recall task is ultimately due to interference, then the form of the stimuli in the list may have differential effects on acceptability. For instance, asking participants to memorize sequences of non-linguistic stimuli, such as numbers or shapes, may be less likely to improve ratings for gapless filler dependencies. If so, this may suggest that linguistic interference underlies the improved acceptability of Memory Load in these studies. Similarly, one way to directly isolate decay as a factor would be to keep the linguistic form of the sentences the same, but modulate the stimulus onset synchrony (SOA), such that there is increased time between the filler NP and the gapless predicate. If increasing the length of time improves the ratings of –Gap predicates, then this demonstrates that decay of the representation over time uniquely contributes to the loss of the predicted structure in memory. Finally, it may be possible to directly modulate the effect of interference on the prediction maintained in memory while keeping sentence length the same by selecting different island types, given an independent measure of syntactic complexity. Alternatively, interference may be increased without affecting sentence length by manipulating the degree of feature overlap between the filler NP and the subsequent linguistic material, to induce similarity-based interference.
In this paper, I provided a mechanistic account of resumption in English. I argued that resumption in English is a complex phenomenon that relies on anaphoric processing and failure to execute typical filler-gap dependency processing. Shortly upon detecting the filler, comprehenders construct a prediction for gapped structure. However, as further material is processed over time, this representation becomes more difficult to access. This results in loss of the representation, which in turn means that comprehenders are less likely to notice if the prediction goes unsatisfied. Because no filler-gap dependency is computed, a reference dependency between the pronoun and filler NP allows the comprehender to relate the filler NP to the meaning of the sentence. This allows the comprehender to build a a coherent interpretation of the ungrammatical sentence. In four studies, I compared the resumptive pronouns with anaphoric pronouns, and resumptive dependencies resolving in islands with later filler-gap dependencies. Across experiments, I manipulating the length of the dependency, and increased strain on memory with the addition of the memory recall task. I showed that increased demand on memory resources decreased sensitivity to whether a filler dependency resolves. Moreover, I showed that coreference between the filler NP and the pronoun improves ratings in a subset of these cases. Additionally, I argued that this proposal is consistent with the generalization that comprehenders deploy grammatical constraints rapidly and effectively in typical sentence-processing, because resumptive dependencies are constructed only when typical, grammatically-constrained processing fails.
The additional file for this article can be found as follows:
‘Materials’. DOI:
F = (stereotypically) feminine, M = (stereotypically) masculine
The studies reported here were approved by the University of Minnesota IRB, Study # STUDY00001444.
The structure of this model using
The structure of this model using
The structure of this model using
I thank an anonymous reviewer for pointing this out to me.
The structure of this model using
In
Interestingly, the meta-analysis in Section 6 demonstrated that longer dependencies overall lowered ratings for +Gap conditions, which may be interpreted as consistent with this proposal. However, it may also reflect an overall trend towards noisier responses.
I would like to thank Colin Phillips, who I developed an earlier version of the proposal in this paper with. Additionally, I would like to thank Aya Meltzer-Asscher, Maayan Keshev, Jason Overfelt, Claire Halpert, Sashank Varma, Tim Hunter, Brian Dillon, Elaine Francis, and the Minnesota Syntax/Psycholinguistics Lab (MSPLab) for helpful comments and discussion. I would also like to thank the audiences of WCCFL 36 and CUNY 31 for feedback.
The author has no competing interests to declare.